Coercive Control and Its Impact on Intimate Partner Violence Through the Lens of an EU-Wide Survey on Violence Against Women

Abstract

Research on intimate partner violence has suggested that not all violence is the same. This article builds upon earlier research on coercive control—or intimate terrorism—and examines the experiences of women who can be assessed as having experienced coercive controlling violence both in terms of the types of violent incidents they experience and the impact and consequences of the most serious incident of violence by an intimate partner. The article explores differences across the 28 European Union (EU) Member States in terms of coercive control and type of violence used. The results—based on data from the first EU–wide survey on violence against women by the EU Agency for Fundamental Rights—are further considered in terms of their association with gender equality. The results show that, in the EU, violence against women perpetrated under coercive control differs from other forms of violence as it involves more serious forms of violence and has a bigger impact in terms of its varied consequences. Countries where women indicated lower levels of coercive control are shown as scoring higher on a measure of gender equality, in contrast with earlier interpretations of the survey findings concerning the relationship between survey measures of physical and/or sexual violence and gender equality. The analysis supports the need to differentiate between various types of intimate partner violence against women—including violence under coercive control—in the European context, both in terms of research to better understand violence and for interventions to prevent violence.

Keywords

domestic violence domestic violence and cultural contexts violence exposure

Introduction

Policy makers, academics, and activists have repeatedly called for better data on violence against women as a tool for evidence-based policies to prevent violence (including revictimization), prosecute the offenders, and protect the victims. At the same time, researchers have sought to interpret the available data to better understand violence against women and intimate partner violence in terms of the type of perpetrator, history of violence, and its consequences. This work has involved the development of typologies of violence that can contribute to services and interventions that are better at meeting the needs and expectations of victims and targeting the available resources.

In particular, some researchers have highlighted the need to differentiate between various forms of intimate partner violence in terms of types of violence used and the impact that the violence has on the victim (M. P. Johnson, 2008; Stark, 2007). In part, this debate has been brought about by the seeming contrast between the high volume of intimate partner violence that has been uncovered through population-based surveys, and the day-to-day reality of victim support services that encounter victims of, predominantly, severe abuse—albeit in lesser numbers than might be expected based on survey estimates. Similarly, as with women’s under-reporting of abuse to victim support, surveys that indicate a high prevalence of violence against women also suggest that relatively few victims go on to report the incident(s) of violence to the police, which is reflected in the number of intimate partner violence incidents recorded by the police. The need to differentiate between various types of intimate partner violence has also been raised by researchers who have found evidence of gender symmetry or asymmetry in intimate partner violence, or who have rejected the relevance of gender (a)symmetry, focusing on other characteristics of intimate partner violence incidents. Recognizing the differences between types of intimate partner violence in terms of consequences of violence can have an impact for further adjustment and improvement of various tools designed for early detection of intimate partner violence and risk assessment.

This article examines the prevalence and consequences of coercive control across 28 countries—the member states of the European Union (EU). The term coercive control (or intimate terrorism) has been introduced in the research literature to refer to a “course of conduct” in intimate partner violence, which according to Stark (2012) leads to “ . . . a hostage-like condition of entrapment that arises from the suppression of a victim’s autonomy, rights and liberties . . . ” (p. 5). In the United Kingdom, the Serious Crime Act (2015) criminalized “controlling or coercive behaviour in an intimate or family relationship,” in response to concerns that earlier criminal law provisions failed to adequately recognize “non-physical” abuse, despite repeated emotional and psychological abuse being harmful to the victim (Home Office, 2012). While the Council of Europe Convention on Preventing and Combating Violence Against Women and Domestic Violence (known as the Istanbul Convention; Council of Europe, 2011a) does not specifically recognize coercive control, it requires state parties to the convention to pay due attention to various forms of psychological violence, alongside physical and sexual violence. Article 33 of the Istanbul Convention requires state parties to take measures to ensure that intentionally impairing a person’s psychological integrity in a serious way is criminalized, and Article 46 requires that both repeated offenses and causing severe psychological harm are considered aggravating circumstances in sentencing, among other issues. In the absence of a separate article in the Convention concerning it, both Articles 33 and 46 of the Istanbul Convention are relevant when considering coercive control as a repeated or continuous form of psychological abuse.

This article uses data from the EU-wide survey on violence against women which the European Union Agency for Fundamental Rights (FRA) carried out in 2012 (FRA, 2014a)—see article by Goodey in this special issue. The survey, which interviewed 42,000 women, was designed to provide for the first time comparable data for the EU concerning the prevalence, nature, and consequences of violence against women. In addition to FRA’s analysis, the data from the survey is available to researchers and further analysis of the results can contribute to better understanding of various forms of violence against women from a multi-country perspective. Importantly, the survey results—providing the first EU-wide overview of the situation—can inform policy measures that the EU and its Member States take when implementing the EU Victims’ Rights Directive (2012/29/EU), which the EU Member States were required to put in practice by November 2015 (EU, 2012). The Directive sets out rights for victims of crime and responsibilities for the EU Member States, including the need to carry out individual assessments of victims and their protection needs, where victims of gender-based violence, violence in a close relationship, and sexual violence should be given particular attention. The Directive also encourages that individual assessments take into account the personal characteristics of victims such as age, ethnicity, sexual orientation, gender identity, disability, or dependence on the perpetrator. At the EU level, the results can also be considered in the implementation of the EU Directive and Regulation, which jointly ensure mutual recognition of emergency barring orders across EU Member States in civil and criminal matters (EU, 2011, 2013). In addition to referring to physical and sexual violence, preambles of both instruments (the Directive and the Regulation) note that they cover protection measures in cases where a person’s psychological integrity is in danger.

In parallel, as countries in Europe are in different stages of signing, ratifying, and implementing the 2012 Istanbul Convention by the Council of Europe, results from the 28-country survey in the EU can support and encourage countries to sign and ratify the Convention as well as help ensure its effective implementation (Council of Europe, 2011a, 2011b). State parties to the Convention are urged also to carry out awareness-raising activities focusing on the general public, education, and training activities; set up preventive intervention and treatment programs; and collect statistics on various forms of violence against women. These developments further increase the need for information concerning the experiences of women who are victims of violence and their needs in terms of medical, social, legal, and other support.

Data from the FRA survey are used in this article to assess the extent of coercive control in women’s experiences of intimate partner violence and the consequences and impact of violence in relationships where coercive control is present. FRA survey data are used to construct a measure of coercive control as severe, repeated psychological abuse and control by the victim’s intimate partner. In particular, the article seeks to determine how the FRA survey results reflect women’s experiences of coercive control across different countries in the EU, what characterizes intimate partner violence where the partner exerts coercive control, and how the presence of coercive control manifests itself in terms of consequences the violence has to the victim. The article concludes with a discussion concerning the use of a (multi-country) survey in measuring coercive control and validating existing theories concerning it. Before FRA’s harmonized multi-country survey, it has not been possible to explore at the macro level in Europe—based on data across 28 countries—the theories suggesting that some forms of intimate partner violence (e.g., so-called situational couple violence) are not as deeply rooted in gender inequality as some other forms of intimate partner violence such as intimate terrorism or coercive control (Jasinski, Blumenstein, & Morgan, 2014).

As the survey interviewed women and not men, the results do not contribute to the discussion concerning gender symmetry or asymmetry in experiencing or perpetrating intimate partner violence, where further research is needed—in particular concerning the presence or absence of gender symmetry across different countries and, specifically, in terms of coercive control (as examined by Jasinski et al., 2014). While the survey collected information about respondents’ characteristics including age, education, ethnicity, disability, and sexual orientation, and the impact of these on the prevalence of various forms of violence against women has been discussed by FRA (2014a), the limitations in the survey’s sample size available at the country level—averaging 1,500 interviews per country—make it unfeasible to extend the current analysis to include these factors. Other researchers have looked at coercive control with respect to different partnerships; for example, Frankland and Brown (2014) have found coercive control also in same-sex relationships based on a nonrepresentative sample of men and women. This area is likely to receive increased attention in the coming years as more research is undertaken on lesbian, gay, bisexual, and transgender (LGBT) relationships.

Coercive Control in Research Literature

Survey findings suggesting that women and men use violence in relationships in more or less equal measure (Kimmel, 2002) led researchers to examine more closely the various characteristics of intimate partner violence. Going beyond prevalence of violence, research has examined the characteristics of the victim, the perpetrator, their relationship, and the situational features to be able to identify risk factors of offending and victimization that could be used to inform policy and practice. Furthermore, researchers have sought to develop different typologies of intimate partner violence to differentiate between forms of violence where gender symmetry exists, and forms of violence where the existence of gender symmetry can be rejected. These typologies have sought to classify incidents of intimate partner violence into separate categories that would help better understand the results concerning gender (a)symmetry as well as the way in which data collected from different sources and by different means (prevalence surveys, administrative statistics) reflect different facets of intimate partner violence. Langhinrichsen-Rohling (2010) and H. Johnson (2015) provide an overview of efforts to construct different typologies and how they have developed over time.

As a part of efforts to understand available research evidence—including data from prevalence surveys and administrative sources—and their implications for practitioners and policy measures, in the 1980s, researchers in the United States identified control asserted by a woman’s partner as a commonly occurring feature in the interviews with women victims of violence who had sought help in women’s shelters (Stark, 2010). Later on, the role of controlling behaviors was integrated as a part of typologies that differentiate between various forms of intimate partner violence, starting from Johnson (1995) who referred to “patriarchal terrorism . . . that involves the systematic use of not only violence, but economic subordination, threats, isolation and other control tactics” (p. 284). In his 2006 paper, M. P. Johnson referred to this as “intimate terrorism,” distinguishing between partner violence that could be characterized as intimate terrorism, situational couple violence, violent resistance, and mutual violent control. Johnson derived these four categories based on his classification of the behavior of a violent partner and the behavior of the other partner who may or may not be violent. In both “situational couple violence” and “intimate terrorism,” one of the partners is violent while the other is either nonviolent or is also using violence—the difference between these two categories is the exercise of (coercive) control by the violent partner which results in “intimate terrorism.” If both partners are violent and controlling, Johnson classifies this as “mutual violent control,” while a situation where the violence by a noncontrolling partner is met with violence and control is defined as “violent resistance.” Overall, in Johnson’s typology, coercive control is an important identifying feature that helps differentiate between the different types of violence, in terms of absence of coercive control or presence of it.

The use of the term coercive control is hampered by the lack of an agreed definition and established ways of measuring it. While Stark (2010) uses it as an umbrella term for violence, intimidation, isolation, and control in a relationship, Crossman, Hardesty, and Raffaelli (2016) and Dutton, Goodman, & Schmidt (2005) note that coercive control can exist without the use of physical or sexual violence. Dutton, Goodman, & Schmidt proposed a detailed model of coercive control which puts in the center the communication of coercive threats by the controlling partner and appraisal of the threat by the victim. Prior incidents of intimate partner violence can serve to prime the victim, so that the controlling partner can exert control through coercive threats, without the need to resort every time to physical and/or sexual violence. Dutton, Goodman, & Schmidt have applied their model to develop detailed scales for measuring coercive control in the form of demands, coercion, and surveillance.

As noted by H. Johnson (2015), while many surveys on intimate partner violence include items that could be analyzed in the framework of coercive control, this is often not done and instead the questions concerning psychological abuse and control are analyzed separately and not in combination with data on physical and sexual violence. Furthermore, Myhill (2015) notes that coercive control has often been studied using small (convenience) samples. Myhill’s research used data from the 2008/2009 round of the Crime Survey for England and Wales (at the time still called the British Crime Survey) to construct a measure of coercive control. Respondents were considered as having experienced coercive control if their partner had both “repeatedly belittled you to the extent that you felt worthless” and “frightened you, by threatening to hurt you or someone close to you.” In Myhill’s study, survey respondents who had experienced other forms of psychological abuse or physical violence were classified as victims of “situational violence” (earlier, M. P. Johnson & Leone, 2005 also defined “situational violence” as intimate partner violence that did not involve coercive control). At the same time, this measure of coercive control provides for a limited coverage of the three domains (demands, coercion, and surveillance) as suggested by Dutton, Goodman, & Schmidt (2005).

Research on the Impact and Consequences of Violence

According to M. P. Johnson (2008), coercive control is a long-term process which has implications on victims’ physical and psychological health as well as their economic well-being. To examine the impact of coercive control in intimate partner violence, it is necessary to understand how consequences of violence are defined and measured. Since the World Health Organization’s (WHO) 2002 World report on violence and health (Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002), the health consequences of interpersonal violence have received increasing attention, leading to WHO launching a Global Campaign for Violence Prevention for the period 2012-2020 (WHO, 2012). Other studies have assessed the impact of violence also beyond health consequences. Notable among these are studies concerning the costs of violence against women that offer a more comprehensive overview of the various types of impact (Centers for Disease Control and Prevention, 2003; Council of Europe, 2012; Day, McKenna, & Bowlus, 2005; European Institute for Gender Equality, 2014; Zhang, Hoddenbagh, McDonald, & Scrim, 2012). Monetary costs have been attached to a variety of consequences of violence, such as lost economic output of victims; use of different services such as legal services, health care services, criminal justice interventions, and victim support services; out-of-pocket legal and health care costs; and costs to the victim in the form of health loss, pain and suffering, and loss of quality of life.

Survey data on the impact of violence form an important evidence base when estimating the costs of violence. Interest in the costs of violence against women—often stemming from the need to present the impact of violence against women to policy makers in tangible, monetary terms—has created demand for more detailed information concerning the different forms of impact and their occurrence. Although studies on costs of violence aim to arrive at an estimate of the monetary costs, some types of impact that these studies take into account may not manifest themselves to the victim as a monetary loss but as a loss of (mental and physical) health, quality of live, feeling of security, and well-being. In the United States, both M. P. Johnson and Leone (2005) and Jasinski et al. (2014) have used the National Violence Against Women Survey to examine the links between intimate terrorism and consequences of violence.

Alternative ways to assess the impact or severity of crime—violent crime as well as other offenses—can involve the use of existing administrative data. An example of this approach is the Crime Severity Index developed by Statistics Canada. This Index can be used to assess the changes in the severity of crime over time, as well as to detect differences in crime profiles across the regions in Canada (Statistics Canada, 2009). The Crime Severity Index takes the police-reported crime statistics and weights the figures based on sentencing data from the court system, so that crimes which are more likely to lead to long prison sentences are given a larger weight in the index, compared with other crimes which do not lead to prison sentences or where the mean prison term is shorter. While data on sentencing are available in most countries and could be used for the purpose of assessing severity of violence, using this approach to assess the severity of violence against women could lead to a skewed picture of reality because only a fraction of incidents are reported to and recorded by the police. Country comparisons based on administrative data are further hampered by the lack of harmonized criminal law definitions of offenses across countries, as well as different sentencing practices. While ongoing initiatives such as the International Classification of Crime for Statistical Purposes—developed by the United Nations (United Nations Office on Drugs and Crime, 2015)—may in the future assist comparisons of police recorded crime, the problem remains that relatively few incidents of violence against women are reported to the police.

Victimization surveys offer a way to complement statistics on police-recorded incidents in two important ways. First, by collecting data from both nonvictims and victims to measure the prevalence of victimization. Second, by asking all those who report their victimization experiences in a survey detailed questions about these incidents. This information can contribute to efforts to prevent victimization and improve service delivery to meet the needs of the victims. Focusing on a single incident or incidents—in victimization surveys often “the most recent incident” or a certain number of incidents in a given time period, or “the most serious incident”—gives the possibility to collect more detailed information concerning the victim’s experiences as well as consequences of violence. The information concerning consequences can be vital for policy makers and service providers who seek to assess how many women have used and are in the need of particular forms of support, such as health care services, housing support and social security, and criminal justice. Information concerning victims’ satisfaction with the assistance they have received and reasons for not seeking various forms of support can help policy makers and professionals working with victims to improve existing services, develop new forms of support, and inform awareness-raising activities to make people better aware of their rights and the support that is available to victims.

In measuring the consequences of violence to the victim, the FRA survey focused on the most serious incident experienced by women since the age of 15—separately concerning violence where the perpetrator was the current partner, a previous partner, or somebody else (nonpartner). Respondents who had experienced only one incident of violence by a given perpetrator were asked automatically to consider this incident, while women who had two or more victimization experiences by the same perpetrator were asked to select the one which had had the biggest impact on them, either physically or psychologically. The questions concerning the prevalence of violence as well as details concerning the most serious incident were asked separately for violence by a partner and violence by a nonpartner (i.e., any other perpetrator than a victim’s partner, such as an acquaintance, relative, colleague, or a stranger). The data available for analysis concerning the impact of violence reflect the experiences of those women who disclosed in the survey what had happened in the most serious incident of physical and/or sexual violence.

Measuring Coercive Control Through the FRA Survey

In 2012, FRA carried out the first EU-wide survey on violence against women, interviewing 42,000 women. FRA developed the survey questionnaire, which was translated into the official languages of the 28 EU Member States, and which asked women about their experiences of physical, sexual, and psychological abuse and control; sexual harassment; stalking; and experiences of violence in childhood. In each country, the interviewees were selected following strict probability sampling procedures, ensuring representativeness of the survey results at the Member State and EU level (for more background information concerning the FRA survey, see article by Goodey in this special issue; also FRA, 2014a, 2014b, 2014c).

FRA did not set out to measure coercive control and, therefore, has not replicated any of the proposed scales (see, for example, Beck, Menke, O’Hara Brewster, & Figueredo, 2009; Dutton, Goodman, & Schmidt, 2005; Graham-Kevan & Archer, 2003). To use the survey data to examine coercive control, it is necessary to assess the questions that have been asked from the perspective of coercive control, drawing upon the various ways coercive control has been defined. However, what seems common to most definitions is the view that coercive control involves acts such as demands, coercion, surveillance (Dutton, Goodman, & Schmidt, 2005); intimidation, isolation, control (Stark, 2010); and economic control, threats, emotional intimidation, abuse, and isolation (Graham-Kevan & Archer, 2003). In the way of clarification, Kelly and M. P. Johnson (2008) use the term “coercive controlling violence” to refer to physical and/or sexual intimate partner violence that takes place in the presence of coercive control. In this article, “coercive control” is used to refer to repeated psychological abuse and control by an intimate partner, and “coercive controlling violence” refers to physical and/or sexual violence in a relationship where coercive control is used.

Compared with the British Crime Survey analysis by Myhill (2015), the FRA survey on violence against women (N = 42,000) included a more extensive set of 17 questions concerning psychological abuse and control (Table 1)—for example, whether the partner tries to keep the woman from seeing her friends or working outside the home, or whether the partner humiliates the woman in public or threatens to harm their children. In its analysis of the survey results, FRA uses the 17 questions on psychological abuse and control to differentiate between “controlling behavior,” “economic violence,” “abusive behavior,” and “blackmail with/abuse of children” (FRA, 2014a).

Table 1.

Forms of Psychological Violence by the Current Partner Asked About in the FRA Survey (FRA, 2014b).

Question E01: When two people are in a relationship, they usually share both good and bad moments. These questions are about your current partner. How often does your current partner . . .

a. try to keep you from seeing your friends?
b. try to restrict your contact with your family of birth or relatives?
c. insist on knowing where you are in a way that goes beyond general concern?
d. get angry if you speak with another man/woman?^a
e. become suspicious that you are unfaithful?
f. prevent you from making decisions about family finances and from shopping independently?
g. forbid you to work outside the home?
h. forbid you to leave the house, takes away your car keys, or locks you up?

Question E02: How often would you say that your current partner has . . .

a. belittled or humiliated you in front of other people?
b. belittled or humiliated you in private?
c. done things to scare or intimidate you on purpose, for example, by yelling and smashing things?
d. made you watch or look at pornographic material against your wishes?
e. threatened to take the children away from you?^b
f. threatened to hurt your children?^b
g. hurt the children?^b
h. threatened to hurt or kill someone else you care about?

Question E03: How often has something like this happened to you? Your current partner has . . .
a. threatened to hurt you physically?

Note. FRA = European Union Agency for Fundamental Rights.

Interviewers were instructed to read out “men” or “women” in accordance with the gender of respondent’s partner, which was recorded earlier in the interview.

Questions concerning children were only asked to respondents who earlier in the interview indicated having children.

For each of the 17 forms of psychological abuse and control, women could indicate whether they had experienced them “never,” “sometimes,” “often,” or “all the time” in their relationship. The survey asked women separately about their current relationship and previous relationships—however, due to the possibility that a respondent had several different partners in the past, it was not possible to ask how often each relationship exhibited the listed behavior, because this would have become very burdensome and repetitive for the respondent to answer. Instead, in the case of previous partners, women were asked whether any of their previous partners had acted in the way described in the survey questions. However, because the frequency with which psychological abuse and control was used is only available for current partner violence, the analysis that follows focuses on women’s experiences with their current partner.

Among the 17 items of psychological abuse and control that were included in the FRA survey (Table 1), it is possible to identify items that address the three dimensions of coercion proposed by Dutton, Goodman, & Schmidt. (2005)—demands, coercion, and surveillance. Dutton, Goodman, & Schmidt further differentiate within the dimension of demands nine subscales: (a) Personal Activities/Appearance, (b) Support/Social Life/Family, (c) Household, (d) Work/Economic/Resources, (e) Health, (f) Intimate Relationship, (g) Legal, (h) Immigration, and (i) Children/Parenting. The dimension of coercion is divided into three subscales: (a) Harm to You, (b) Harm to Partner, and (c) Harm to Others; while the dimension of surveillance is not split into subscales. Questions included in the FRA survey address five of the nine subscales of dimension of demands—the questions asked in the survey concerning psychological abuse and control do not address coercive control in terms of subscales that have to do with the household (such as taking care of the house), health (such as controlling when the respondent can drink alcohol), legal matters (such as forcing a woman to do things that are against the law), or immigration (such as threats to withdraw support for a citizenship application). Out of three subscales of coercion, FRA’s survey asked questions related to the perpetrator harming the respondent or harming others (such as the children), but without asking whether the partner threatened to harm himself as a form of exerting control. Finally, the survey’s questions concerning psychological abuse and control included questions on surveillance as a way for the perpetrator to control the partner. The items on psychological abuse and control included in the FRA survey have been selected based on their use in other established multicountry surveys such as the International Violence Against Women Survey (IVAWS; H. Johnson, Ollus, & Nevala, 2008) and the WHO’s (2005) multicountry survey.

For each of the items concerning psychological abuse and control, women could indicate whether they had experienced their current partner doing this “never,” “sometimes,” “often,” or “all the time.” Based on women’s responses concerning their current relationship three categories can be formed: (a) women who have never experienced any of the listed forms of psychological abuse and control in their current relationship, (b) women who indicated that they sometimes experienced one or more of the listed forms of psychological abuse and control in their current relationship, but not more often than that, and (c) women who indicated in the survey that one or more of the listed forms of psychological abuse and control takes place in their current relationship “often” or “all the time.” In the analysis that follows, these groups are referred to as experiencing “none,” “moderate,” or “high” levels of psychological abuse or control by their current partner, and women who indicated having experienced high levels of psychological abuse or control (incidents occurring “often” or “all the time”) are referred to as experiencing coercive control.

Defining coercive control in terms of the frequency of psychological abuse and control has the shortcoming that it does not take into account the impact that this abuse and control has, and the fact that a single incident may have a considerable impact on the victim (when the perceived consequences are tangible and severe). However, coercive control is usually defined as a repeated or continuous conduct, and by measuring victims’ experiences over the life course (in the FRA survey, since the age of 15 for most forms of violence), it is possible to capture ongoing control and experiences of other forms of violence—for example, where the offender has used physical violence sometime in the past, followed by coercive control where the threat of violence continues to be present. Nevertheless, the measure of coercive control that is used in this article would not capture possible cases of coercive control where—according to the victim—the offender engages in one or more of the 17 forms of psychological abuse and control “sometimes,” but none of them “often” or “all the time,” and therefore the actual prevalence of coercive control may be higher. A further consideration is that the FRA survey data cannot be used to differentiate between cases where coercive control is mutual, as opposed to relationships where only one of the partners uses coercive control, because the survey focused on women’s experiences in the relationship.

Coercive Control Across the 28 EU Member States

Figure 1 shows the prevalence of psychological abuse and control in the 28 EU Member States and the EU average. The vertical bars are further subdivided into women who have experienced high levels of (coercive) control (i.e., who indicated that they experience one or more of the controlling behaviors “often” or “all the time”—bottom section of the bars) and women who sometimes experience psychological abuse and control in their current relationship, but this does not happen in their opinion “often” or “all the time” (top section of the bars).

Figure 1.

Controlling behavior by the current partner by level of control in the FRA survey, EU-28 (%).

The percentage of women who experience high levels of coercive control does not follow in a linear way the percentage of women who sometimes experience psychological abuse and control. In particular, in some countries where the overall rate of psychological abuse and control is high—such as Denmark, Estonia, Finland, and the Netherlands—the prevalence of women experiencing coercive control is relatively low, compared with the percentage of women experiencing coercive control in other countries. In the survey, the highest prevalence of coercive control was found in Lithuania (16%), Bulgaria (11%), Hungary (11%), Latvia (10%), and Romania (10%). The countries with the lowest prevalence of coercive control by a woman’s current partner are Sweden (2%), United Kingdom (3%), Ireland (3%), Denmark (4%), and the Czech Republic (4%). At the country level, the prevalence of coercive control is positively correlated with rates of physical and/or sexual violence by an intimate partner (Figure 2).

Figure 2.

Physical and/or sexual violence and women experiencing a high level of (coercive) control by the current partner in the FRA survey, EU-28 (%).

A result of the FRA survey which has received considerable attention concerns the positive correlation between the prevalence of physical and/or sexual violence since the age of 15 and scores in the EU Gender Equality Index (European Institute for Gender Equality, 2015). FRA’s initial analysis suggested that rates of physical and/or sexual violence against women are higher in countries which score high in terms of gender equality, although the positive impact of gender equality could be posited to contribute to lower levels of intimate partner violence.

It is therefore notable that examining coercive control against countries’ scores on the gender equality index of the European Institute for Gender Equality (EIGE) produces a starkly different pattern—that is, in countries where the level of gender equality is estimated as high, women in the FRA survey indicated relatively low levels of coercive control, r(26) = −.60, p < .01 (Figure 3). An interpretation of these results suggests that levels of coercive control are more closely related to gender equality than rates of physical and/or sexual violence against women overall, as measured in the FRA survey. This can support the view that asymmetrical gender dynamics produce—directly or indirectly—coercive control, while prevalence of physical and/or sexual violence in a country may be governed primarily by other factors. Accompanying its overview of survey findings, FRA (2014a) presented a variety of factors, which may explain differences between countries in terms of prevalence of violence against women. These factors include social acceptability of speaking openly with other people about experiences of violence against women, gender equality contributing to higher levels of disclosure of violence, women’s exposure to violence increasing if more women work outside the home, overall levels of violent crime, and differences in drinking patterns. However, the finding suggesting that countries with high scores on gender equality have low prevalence of coercive control adds a new dimension to the discussion concerning the prevalence of violence and gender equality. Namely, specific types of intimate partner violence (such as coercive controlling violence) may have different correlates which cannot be accurately identified based on an analysis that looks at the overall prevalence of intimate partner violence without differentiating between various types of violence.

Figure 3.

Scores on EIGE’s Gender Equality Index 2012 by prevalence of high levels of (coercive) control (%) in the FRA survey.

Types of Violence Experienced as a Measure of Impact

In the FRA survey, experiences of physical violence were approached through a set of questions which focused on nine acts of violence (Table 2). The same list of acts of physical violence was separately applied for violent acts where the perpetrator was a partner—either current or previous—or nonpartner (i.e., any other person than the partner). For each of the nine violent acts, respondents could indicate whether they had experienced them since the age of 15—and if yes, whether they had taken place in the 12 months before the interview. Women could indicate all acts that they had experienced, be it in a single incident or several incidents over time on different occasions.

Table 2.

Forms of Physical Violence Asked About in the FRA Survey (FRA, 2014b).

Since you were 15 years old until now/in the past 12 months, how often has someone

a. pushed you or shoved you?
b. slapped you?
c. thrown a hard object at you?
d. grabbed you or pulled your hair?
e. beaten you with a fist or a hard object, or kicked you?
f. burned you?
g. tried to suffocate or strangle you?
h. cut or stabbed you, or shot at you?
i. beaten your head against something?

Note. FRA = European Union Agency for Fundamental Rights.

While each of the nine acts of physical violence listed in the survey can include incidents of varying severity—and FRA’s analysis of the survey results did not differentiate between them in terms of their seriousness—some of the listed acts of physical violence can be seen as more likely to lead to serious injuries or to seriously frighten or otherwise result in a negative psychological impact for the victim. For a more detailed analysis of the violent acts experienced, the list in Table 2 can be divided into two groups, where Items a to d represent acts of violence which are less likely to lead to severe physical injuries and/or psychological impact, compared with acts described in Items e to i. It is important to note, however, that the nine acts of violence—as listed in the survey—can only provide a proxy measure of impact, as each of them may lead to a wide range of consequences for the victim, and even those acts classified as “less severe” such as being pushed or shoved may in some instances lead to serious bodily harm.

The following analysis examines the types of violence experienced based on respondents who indicated having experienced one or more types of physical violence by a current or previous partner, out of the nine types of physical violence listed in Table 2. Due to the low prevalence of some forms of serious physical violence, it would not have been possible to focus only on incidents of serious physical violence where the perpetrator has been the current partner due to the limitations posed by the sample size available for analysis at the EU Member State level—in contrast to the measure of coercive control that was presented earlier in this article where the analysis focused on all incidents (serious as well as less serious) of physical violence perpetrated by the current partner.

Figure 4 shows the percentage of women who have experienced forms of violence classified as less severe (Items a-d) as well as the percentage of women who have experienced one or more of the severe forms of violence (Items e-i), by country. Given that, in the survey, the questions concerning experiences of violence were asked both concerning the current partner and (any) previous partners, it is possible to examine these results in terms of women’s experiences of violence by any partner (i.e., current or previous). On average, using this measure of severity, 12% of women in the EU have experienced less severe physical violence by their partner, while 8% of women have experienced severe physical violence by a partner since the age of 15.

Figure 4.

Less severe and more severe physical violence by any partner (current and/or previous) in the FRA survey, EU-28 (%).

Comparing the percentage of respondents who experienced less severe forms of violence and the percentage of those who experienced more severe forms of violence suggests that in almost all EU Member States, women were more likely to indicate in the survey that they had experienced one or more of the acts described in Items a to d (less severe), compared with experiencing acts e to i (more severe). Exceptions from this are Ireland and United Kingdom, where the percentage of women who indicated having experienced more serious forms of physical violence (Items e-i) is higher than the percentage of women who experienced less serious forms of violence (Items a-d).

The type of violence experienced as a measure of impact can be further explored by comparing it with the measures of psychological abuse and control, as discussed earlier in this article. Table 3 shows that the relationship between respondents’ experiences of psychological abuse and control, and physical violence by their partners is pronounced. Women who indicated in the survey that their partner has never exhibited the listed forms of controlling behavior in the relationship had rarely experienced physical violence by the said partner—1% had experienced one or more of the less severe forms of physical violence, and close to 0% had experienced severe forms of physical violence. However, out of women who have experienced high levels of psychological abuse and control (i.e., coercive control), 32% had experienced less severe forms of physical violence, and 23% had experienced severe forms of physical violence.

Table 3.

Percentage of Women Experiencing Physical Violence by the Current Partner in the FRA Survey, by Intensity of Violence and Partner’s Controlling Behavior.

	Intensity of Physical Violence
	No Violence	Less Severe Violence	More Severe Violence	Total
Intensity of control by the current partner
None	99	1	0	100
Moderate	84	14	2	100
High	45	32	23	100
Total	93	5	2	100

Source. FRA Violence Against Women Survey dataset, 2012.

Note. FRA = European Union Agency for Fundamental Rights.

It was noted earlier based on FRA (2014a) research that, at the country level, the prevalence of physical and/or sexual violence is positively correlated with EU Member States’ scores on the EIGE Gender Equality Index. The same is true for prevalence of serious forms of physical violence, r(26) = .34, ns, and less serious physical violence, r(26) = .26, ns, by a partner, and that both correlations are positive suggests that countries with higher values on the EIGE index also have a high prevalence of both types of physical violence by a partner—at the same time, both correlations are statistically not significant due to the small number of observations—in this case, 28 countries. This is in contrast with the results concerning coercive control presented earlier in this article, where the correlation with the Gender Equality Index was stronger as well as negative—that is, a high rate of coercive control was associated with a low gender equality score. These results suggest that, in the 28 countries examined, neither the prevalence of serious forms of intimate partner violence nor the prevalence of less serious forms of intimate partner violence show similar correlations with EIGE’s measure of gender equality than what could be observed in the case of coercive control earlier in this article.

Impact and Consequences of Violence and Coercive Control

Another option for examining the impact of violence through the FRA survey involves the use of information victims provided concerning the most serious incident of violence they had experienced. The National Intimate Partner and Sexual Violence Survey carried out in the United States by the Centers for Disease Control and Prevention (CDC) provides an example of how different survey questions concerning consequences of violence can be used to develop a measure of impact of intimate partner violence (Black et al., 2011). The CDC survey report estimates the percentage of women in the United States who—during their lifetime—have experienced any intimate partner violence, and its related impact, which is defined as having experienced at least one out of 12 consequences, ranging from fearfulness, posttraumatic stress disorder (PTSD) symptoms and physical injuries, through to the need for various support services or missing work or school. The CDC indicates that, based on its nation-wide survey, 29% of women in the United States have experienced some form of intimate partner violence–related impact during their lives. The most common forms of impact include fear (26%), concerns for safety (22%), and PTSD symptoms (22%).

The FRA survey on violence against women also asked detailed questions concerning the most serious incident of violence that women had experienced since the age of 15. While it is not possible to use the FRA survey to reproduce exactly the same categories as those used in the CDC study due to differences in the questionnaires of the two surveys, questions in the FRA survey concerning the impact and consequences of the most serious incident of violence can be examined without making direct comparisons between the two surveys. Reflecting on the set of items put forth in the CDC study, the FRA survey can be used to produce the set of measures shown in Figure 5 for impact and consequences of the most serious incident of violence when the perpetrator was the victim’s current or previous partner. The results in Figure 5 concerning consequences and impact of physical and/or sexual violence by a partner refer to the percentage of victims who indicated that they had either experienced a particular type of impact or they had taken certain action such as contacting the police.

Figure 5.

Consequences and impact of physical and/or sexual violence in the FRA survey, by type of partner (current/previous), percentage of victims describing the most serious incident.

Overall, looking at results at the level of the EU average and taking together all different types of impact shown in Figure 5, 79% of women victims of violence by their current partner said that the most serious incident involved one or more of the impact types listed in Figure 5, while 90% of victims of violence by the previous partner had experienced at least one of the listed forms of impact (because a victim may have experienced more than one type of impact; the overall percentage of victims who have experienced at least one type of impact is not a simple sum of percentages shown in Figure 5). A major contributor to differences in the results obtained by the CDC in the United States and FRA for the EU can be found in the formulation of questions and composition of the measures.

Based on the categories shown in Figure 5, more detailed analysis of women’s experiences can be carried out differentiating between women who have experienced physical violence by their current partner together with coercive control (coercive controlling violence) and women who have experienced physical violence but without—or a lesser degree of—control by their partner.

Table 4 presents the results concerning the most serious incident of physical and/or sexual violence by a woman’s current partner, differentiated based on the presence (or absence) of coercive control in the relationship. The results support the view that incidents where coercive control is present differ from other violence (such as situational violence). Women who indicated that their current partner “often” or “all the time” used some form of control were more likely to indicate in the survey that the most serious incident resulted in them feeling shocked or fearful. These women were also more likely to indicate that they had some psychological consequences and physical injuries from the most serious incident, although there is no difference in terms of need of medical assistance. Women whose current partner had been violent and controlling were more likely to bring the most serious incident to the attention of the police or to contact other services—this may reflect the, overall, more serious nature of coercive controlling violence in terms of injuries and psychological consequences. They also more often felt that they would have needed some form of protection. And it is worth noting, violence in relationships that involved coercive control was also more likely to take place when the woman was pregnant, compared with violence in a relationship where coercive control was less present or not used at all.

Table 4.

Percentage of Women Indicating in the FRA Survey That the Most Serious Incident of Physical and/or Sexual Violence by the Current Partner Had Specific Consequences and Impacts, by Presence of Control by the Current Partner.

Victim . . .	Victims Experiencing Consequences in the Most Serious Incident of Physical Violence (%)
Victim . . .	No Controlling Behaviors in the Relationship	Presence of Controlling Behaviors in the Relationship
Experienced shock or fear	41	63
Experienced any psychological consequences	49	81
Experienced physical injuries	18	46
Needed to take time off work or studies because of the incident	3	8
Was in the need of medical help	3	4
Contacted services (other than police) following the incident	11	30
Incident came to the attention of the police	6	21
Unweighted n ^a	950	1,029

Source. FRA Violence Against Women Survey dataset, 2012.

Note. FRA = European Union Agency for Fundamental Rights.

With the exception of category “Needed to take time off work or studies because of the incident,” where unweighted n is 811 and 792, respectively, for women who have experienced controlling behaviors in their current relationship and women whose current relationship does not involve controlling behaviors. The difference in unweighted n, compared with other categories, is due to excluding women who were not working or studying at the time of the most serious incident (and therefore the question of taking time off was not applicable).

Discussion

The results concerning impact and consequences of violence demonstrate that coercive controlling violence is associated with more severe impact to victims compared with physical intimate partner violence without coercive control, and that—as a concept—coercive control has value in terms of differentiating between different types of violence against women. Violence by a controlling partner is more likely to result in a measurable impact on a range of indicators, from a victim’s emotional reaction through to the need for organized support measures and police intervention. Given that more than half of women whose current partner uses coercive control have experienced physical violence in the relationship, questions concerning psychological abuse and control could be used by professionals in various contexts as a proxy measure indicating the potential for intimate partner violence.

The survey results show that women in EU Member States experience coercive control to different degrees—however, the differences in the extent of coercive control (psychological abuse and control taking place “often” or “all the time’) are not as large as with total prevalence of psychological abuse and control in the FRA survey. Furthermore, the rate of coercive control in a country is negatively correlated with the EIGE measure of gender equality, with higher levels of coercive control in the country associated with low levels of gender equality and vice versa. This finding is in contrast to results reported by FRA, which suggested that rates of physical and/or sexual violence and gender equality are positively correlated—that is, countries with high scores on EIGE’s gender equality index tend to have high prevalence of physical and/or sexual violence, as measured in the FRA survey. This suggests that the existence of coercive control may be more closely related to factors that also produce gender inequality, while prevalence of physical and/or sexual violence is partly determined by other issues. Given that violence perpetrated together with coercive control (coercive controlling violence) in women’s lives accounts only for a part of violent incidents, this multicountry analysis supports the view that other incident types—classified as situational couple violence, violent resistance, and mutual violent control by M. P. Johnson (2006)—may be linked to different correlates, such as those suggested by FRA (2014a) in its analysis of the violence against women survey results and as discussed earlier in this article. Future research on correlates of intimate partner violence should therefore take into account the type of intimate partner violence in question, as correlates may differ between coercive controlling violence and situational couple violence, as an example. The cross-country correlation between the prevalence of coercive control and gender equality supports the view of Stark (2007), M. P. Johnson (2008), and others who have suggested a relationship between gender equality and coercive control. This article provides support to these theories, for the first time based on an analysis of a harmonized 28-country survey, which included a broad set of questions that allow for constructing a measure of coercive control.

At the same time, it is outside the scope of this article to examine violent resistance and mutual control—as outlined by M. P. Johnson—as the FRA survey did not collect information concerning the extent to which women have used violence, either in response to the violence perpetrated by a partner or independently. Therefore, the analysis in this article can differentiate physical violence that has involved coercive control, but—in the absence of survey questions about women’s use of violence and the events leading up to use of violence—it is not possible to break down violence into further subcategories, such as situational couple violence, violent resistance, and mutual violent control. Identifying the dynamics that contribute in various degrees to different types of violence may be helpful for future research on intimate partner violence—that is, if incidents of coercive control are examined separately, it may be easier to identify contributing factors and correlates that help better understand other forms of intimate partner violence and the differences in prevalence rates across the EU Member States.

Survey research on intimate partner violence depends on respondents’ willingness to share their personal experiences which can be sensitive—indeed, the results of the FRA survey show that many victims felt shocked, embarrassed, or even ashamed after the most serious incident of violence that they had experienced since the age of 15. In the FRA survey, at the end of the interviewer-administered face-to-face interviews, respondents were asked to complete a one-page paper questionnaire concerning their experiences, and the answers were sealed in an envelope without the interviewer seeing them. Comparison of the results of the self-completion questionnaires with the information which the respondents provided directly to the interviewer across the 28 countries did not find notable differences between the two approaches. However, some respondents may always feel unable to share their experiences in a survey. Considering the questions about various forms of violence that were covered in the FRA survey, it is possible that respondents find it easier to answer the items that concern psychological abuse and control, compared with questions that deal with physical and sexual violence. FRA also argued that women’s awareness of rights and the extent to which people find it acceptable in a certain country to talk about experiences of violence may have contributed to the fact that the survey found higher rates of physical and/or sexual violence in countries where, over the years, debates about intimate partner violence have entered the mainstream in the media and political arenas. The results presented in this article partly challenge this explanation, at least in terms of specific forms of intimate partner violence such as coercive controlling violence. Countries with high scores on gender equality such as the Nordic countries—which are the countries often associated with progressive agendas in terms of gender equality—were shown to have low prevalence of coercive control, while countries with low scores on gender equality were more likely to have higher prevalence of coercive control. Therefore, at least in terms of coercive control, the results presented in this article do not support the view that FRA’s survey results would simply reflect differences between countries in the degree to which women feel comfortable to share their experiences, and the differences between the countries in terms of rates of various forms of violence require a more thorough analysis.

The results show that coercive control by a partner—by its very nature repeated or constant psychological abuse and control—is often associated with women experiencing various acts of physical violence as well as more serious and wide-ranging consequences as a result of the most serious incident. These results further support the need to develop effective ways of early identification of high-risk situations by focusing on relationship dynamics, in addition to developing better intervention measures given that violence that takes place in the context of coercive control is likely to reoccur. As an example, multi-agency risk assessment conferences (MARACs) have been developed as a way of coordinating service providers’ work in cases where there is a high risk of serious victimization and reoffending, and this approach—first developed in Cardiff, UK (Robinson & Tregidga, 2005)—has been adopted as a promising practice in a number of countries. In MARACs, various agencies working with victims of domestic violence come together to share information concerning cases of violence against women to reduce repeat victimization, among other aims, but because of their limited capacity, MARACs must prioritize the cases they are dealing with using dedicated risk assessment tools. Following the U.K.’s Serious Crime Act (2015), the National MARAC Scrutiny Panel—which is chaired by the Home Office—has issued guidance for MARACs concerning coercive control and how it can be identified in the work of various agencies that work with victims of violence (SafeLives, 2015).

The fact that the role of coercive control is being recognized in the United Kingdom both in law and in practice should serve as a push for other EU Member States to take into account this form of severe psychological abuse and control, the grave effects of which can be demonstrated through the results of FRA’s multi-country survey. Recognizing coercive control more widely in the EU Member States would also contribute to it being taken into account when victims seek emergency barring orders and in the mutual recognition of the barring orders across EU Member States, as set out by the EU (2011, 2013). While the EU Directive and Regulation on mutual recognition of protection orders refer to a danger to person’s psychological integrity, it is unclear to what extent this is recognized in the different EU Member State’s laws that concern intimate partner violence and to what extent victims enjoy the same level of protection against psychological abuse and control across the EU.

Conclusion

The current article examined the extent and consequences of coercive control on women’s experiences of physical partner violence in the EU Member States. Based on the survey data from 28 countries, the results of the analysis confirm that violence perpetrated by an intimate partner who exercises coercive control is more likely to lead to various harmful consequences for the victim, compared with intimate partner violence without coercive control. At the level of EU Member States, survey prevalence rates for coercive control are closely correlated with gender equality in the country, with higher prevalence of coercive control in countries with low scores on gender equality.

Coercive control, as a severe form of psychological abuse and control, in many cases occurs alongside other forms of violence, although this may not be always the case. Therefore, existing risk assessment tools should pay particular attention not only to signs of physical violence but also to signs of coercive control as a severe form of abuse on its own and as a potential precursor of other forms of intimate partner violence. EU Member States can in this respect learn from the work done in the United Kingdom, where coercive control has recently been addressed in the criminal law, and existing risk assessment tools (in the context of MARACs) have been revised to recognize coercive control. While Article 33 of the Council of Europe’s Istanbul Convention requires state parties to take legislative and other measures to criminalize psychological violence, in the implementation of this article, particular attention could be afforded to coercion and control, which has already entered the criminal law in some countries such as the United Kingdom. In a similar manner, the concept of coercive control could be used in the context of the EU Victims’ Rights Directive as a way to identify particularly vulnerable victims of intimate partner violence in the individual assessment of victims (as specified under Article 22 in the Directive). At the same time, criminal law measures may have a limited impact on reducing coercive control, and the results of a macro analysis at the country level as presented in this article suggest that measures toward achieving gender equality—which can require action across a wide range of areas such as employment, education, and social security—may contribute to lower levels of coercive control. Further research on coercive control is needed in terms of the forms coercive control may take across countries, and the extent to which prevalence of coercive control differs across different groups of victims, for example, based on age and sexual orientation.

Footnotes

Appendix

List taken from http://publications.europa.eu/code/en/en-370100.htm, countries in alphabetical order.

Austria	AT
Belgium	BE
Bulgaria	BG
Croatia	HR
Cyprus	CY
Czech Republic	CZ
Denmark	DK
Estonia	EE
Finland	FI
France	FR
Germany	DE
Greece	EL
Hungary	HU
Ireland	IE
Italy	IT
Latvia	LV
Lithuania	LT
Luxembourg	LU
Malta	MT
Netherlands	NL
Poland	PL
Portugal	PT
Romania	RO
Slovakia	SK
Slovenia	SI
Spain	ES
Sweden	SE
United Kingdom	UK

Author’s Note

Researchers are encouraged to work with FRA’s violence against women survey dataset, which can be accessed as follows: https://discover.ukdataservice.ac.uk/catalogue/?sn=7730&type=Data%20catalogue – or by contacting the Fundamental Rights Agency at statistics&surveys@fra.europa.eu.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Disclaimer

The views expressed in this article are solely those of the author and do not necessarily reflect the views or position of the European Union Agency for Fundamental Rights.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biography

Sami Nevala, MSc, is the head of sector for statistics and surveys at the European Union Agency for Fundamental Rights. His research interests include intimate partner violence, survey research methods, and crime statistics.

References

Beck

C. J. A.

Menke

J. M.

O’Hara Brewster

Figueredo

A. J.

(2009). Validation of a measure of intimate partner abuse with couples participating in divorce mediation. Journal of Divorce & Remarriage, 50, 295-308.

Black

M. C.

Basile

K. C.

Breiding

M. J.

Smith

S. G.

Walters

M. L.

Merrick

M.T.

. . . Stevens

M. R.

(2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 summary report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Centers for Disease Control and Prevention. (2003). Costs of intimate partner violence against women in the United States. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Council of Europe. (2011a). Council of Europe convention on preventing and combating violence against women and domestic violence (Council of Europe Treaty Series No. 210). Retrieved from http://www.conventions.coe.int/Treaty/Commun/QueVoulezVous.asp?CL=ENG&NT=210

Council of Europe. (2011b). Explanatory report to the council of Europe convention on preventing and combating violence against women and domestic violence (Council of Europe Treaty Series No. 210). Retrieved from http://rm.coe.int/CoERMPublicCommonSearchServices/DisplayDCTMContent?documentId=09000016800d383a

Council of Europe. (2012). Overview of studies on the costs of violence against women and domestic violence—Updated 14 October 2014. Strasbourg, France: Council of Europe. Retrieved from https://rm.coe.int/CoERMPublicCommonSearchServices/DisplayDCTMContent?documentId=090000168059aa22

Crossman

K. A.

Hardesty

J. L.

Raffaelli

(2016). “He could scare me without laying a hand on me”: Mothers’ experiences of nonviolent coercive control during marriage and after separation. Violence Against Women, 22, 454-473.

Day

McKenna

Bowlus

(2005). The economic costs of violence against women: An evaluation of the literature—Expert brief compiled in preparation for the Secretary-General’s in-depth study on all forms of violence against women. New York, NY: United Nations.

Dutton

M. A.

Goodman

Schmidt

R. J.

(2005). Development and validation of a coercive control measure for intimate partner violence. Washington, DC: National Institute of Justice.

10.

European Institute for Gender Equality. (2014). Estimating the costs of gender-based violence in the European Union. Luxembourg City, Luxembourg: Publications Office of the European Union.

11.

European Institute for Gender Equality. (2015). Gender Equality Index 2015—Measuring gender equality in the European Union 2005-2012. Luxembourg City, Luxembourg: Publications Office of the European Union.

12.

European Union. (2011). Directive 2011/99/EU of the European Parliament and of the Council of 13 December 2011 on the European protection order. Retrieved from http://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1420623225217&uri=CELEX:32011L0099

13.

European Union. (2012). Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012 establishing minimum standards on the rights, support and protection of victims of crime, and replacing Council Framework Decision 2001/220/JHA. Retrieved from http://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1436278653998&uri=CELEX:32012L0029

14.

European Union. (2013). Regulation (EU) No 606/2013 of the European Parliament and of the Council of 12 June 2013 on mutual recognition of protection measures in civil matters. Retrieved from http://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1489095662131&uri=CELEX:32013R0606

15.

European Union Agency for Fundamental Rights. (2014a). Violence against women: An EU-wide survey—Main results. Luxembourg City, Luxembourg: Publications Office of the European Union.

16.

European Union Agency for Fundamental Rights. (2014b). Violence against women: An EU-wide survey—Questionnaire. Retrieved from http://fra.europa.eu/sites/default/files/fra-violence-against-women-survey-questionnaire-1_en.pdf

17.

European Union Agency for Fundamental Rights. (2014c). Violence against women: An EU-wide survey—Technical report. Luxembourg City, Luxembourg: Publications Office of the European Union.

18.

Frankland

Brown

(2014). Coercive control in same-sex intimate partner violence. Journal of Family Violence, 29, 15-22.

19.

Graham-Kevan

Archer

(2003). Intimate terrorism and common couple violence—A test of Johnson’s predictions in four British samples. Journal of Interpersonal Violence, 18, 1247-1270.

20.

Home Office. (2012). Cross-government definition of domestic violence—A consultation: Summary of responses. London, UK: Home Office. Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/157800/domestic-violence-definition.pdf

21.

Jasinski

Blumenstein

Morgan

(2014). Testing Johnson’s typology: Is there gender symmetry in intimate terrorism? Violence and Victims, 29, 73-88.

22.

Johnson

(2015). Degendering violence. Social Politics, 22, 390-410.

23.

Johnson

Ollus

Nevala

(2008). Violence against women: An international perspective. New York, NY: Springer.

24.

Johnson

M. P.

(1995). Patriarchal terrorism and common couple violence: Two forms of violence against women. Journal of Marriage and the Family, 57, 283-294.

25.

Johnson

M. P.

(2006). Conflict and control: Gender symmetry and asymmetry in domestic violence. Violence Against Women, 12, 1003-1018.

26.

Johnson

M. P.

(2008). A typology of domestic violence: Intimate terrorism, violent resistance, and situational couple violence. Boston, MA: Northeastern University Press.

27.

Johnson

M. P.

Leone

J. M.

(2005). The differential effects of intimate terrorism and situational couple violence—Findings from the National Violence Against Women Survey. Journal of family issues, 26, 322-349.

28.

Kimmel

M. S.

(2002). “Gender symmetry” in domestic violence—A substantive and methodological research review. Violence Against Women, 8, 1332-1262.

29.

Krug

E. G.

Dahlberg

L. L.

Mercy

J. A.

Zwi

A. B.

Lozano

(Eds.). (2002). World report on violence and health. Geneva, Switzerland: World Health Organization.

30.

Langhinrichsen-Rohling

(2010). Controversies involving gender and intimate partner violence in the United States. Sex Roles, 62, 179-193.

31.

Myhill

(2015). Measuring coercive control: What can we learn from National Population Surveys? Violence Against Women, 21, 355-375.

32.

Robinson

Tregidga

(2005). Domestic violence MARACs (Multi-Agency Risk Assessment Conferences) for very high-risk victims in Cardiff, Wales: Views from the victims. Cardiff, UK: Cardiff University.

33.

SafeLives. (2015). Guidance for Maracs—Coercive control. Retrieved from http://www.safelives.org.uk/sites/default/files/resources/Coercive%20control%20guidance%20for%20MARACs.pdf

34.

Serious Crime Act 2015. Available from http://www.legislation.gov.uk

35.

Stark

(2007). Coercive control: How men entrap women in personal life. New York, NY: Oxford University Press.

36.

Stark

(2010). Do violent acts equal abuse? Resolving the gender parity/asymmetry dilemma. Sex Roles, 62, 201-211.

37.

Stark

(2012). Re-presenting battered women: Coercive control and the defense of liberty [Manuscript for Rinfret-Raynor

Lesieux

É.

Cousineau

M.-M.

Gauthier

Harper

(2014). Violence envers les femmes: Réalités complexes et nouveaux enjeux dans un monde en transformation. Quebec City, Canada: Presses de l’Université du Quebéc]. Retrieved from http://www.stopvaw.org/uploads/evan_stark_article_final_100812.pdf

38.

Statistics Canada. (2009). Measuring crime in Canada: Introducing the Crime Severity Index and improvements to the Uniform Crime Reporting Survey. Ottawa, Ontario: Statistics Canada.

39.

United Nations Office on Drugs and Crime. (2015). International Classification of Crime for Statistical Purposes (ICCS)—Version 1.0. Vienna, Austria: Author.

40.

World Health Organization. (2005). WHO multi-country study on Women’s Health and Domestic Violence against Women—Initial results on prevalence, health outcomes and women’s responses. Geneva, Switzerland: Author.

41.

World Health Organization. (2012). Global campaign for violence prevention: Plan of action for 2012-2020. Geneva, Switzerland: Author.

42.

Zhang

Hoddenbagh

McDonald

Scrim

(2012). An estimation of the economic impact of spousal violence in Canada, 2009. Ottawa, Ontario, Canada: Department of Justice.